This invention relates to antiviral agents and methods of use thereof. More particularly, the invention relates to antiviral agents that specifically destroy cells infected by viruses that produce a protease in such infected cells. The antiviral agents are activated by the viral protease, thereby specifically targeting the infected cells for destruction. Toxins that target cell surface receptors or antigens on tumor cells have attracted considerable attention for treatment of cancer. E.g., I. Pastan and D. FitzGerald, Recombinant Toxins for Cancer Treatment, 254 Science 1173-1177 (1991); Anderson et al., U.S. Pat. Nos. 5,169,933 and 5,135,736; Thorpe et al., U.S. Pat. No. 5,165,923; Jansen et al., U.S. Pat. No. 4,906,469; Frankel, U.S. Pat. No. 4,962,188; Uhr et al., U.S. Pat. No. 4,792,447; Masuho et al., U.S. Pat. Nos. 4,450,154 and 4,350,626. These agents include a cell-targeting moiety, such as an antigen-binding protein or a growth factor, linked to a plant or bacterial toxin. They kill cells by mechanisms different from conventional chemotherapy, thus potentially reducing or eliminating cross resistance to conventional chemotherapeutic agents.
Ricin and other similar plant toxins, such as abrin, modeccin and viscumin, comprise two polypeptide chains (known as the A and B chains) linked by a disulfide bridge, one chain (the A chain) being primarily responsible for the cytotoxicity and the other chain (the B chain) having sites that enable the molecule to bind to cell surfaces. Such toxins are known as type II ribosome-inactivating proteins or RIPs. F. Stirpe et al., Ribosome-inactivating Proteins from Plants: Present Status and Future Prospects, 10 Biotechnology 405-412 (1992).
Ricin is produced in the plant Ricinus communis (commonly known as the castor bean plant) via a precursor protein known as xe2x80x9cpreproricin.xe2x80x9d Preproricin comprises a single polypeptide chain that includes a leader sequence, the A chain, a linker peptide, and the B chain. The leader sequence is subsequently removed in the organism to yield proricin, which is then cleaved to eliminate the linker region such that the A and B chains remain connected only by a disulfide bond in the mature protein. The toxicity of ricintype toxins operates in three phases: (1) binding to the cell surface via the B chain; (2) penetration of at least the A chain into the cytosol via intracellular organelles, and (3) inhibition of protein synthesis through the A chain cleaving an essential adenine residue from ribosomal RNA. Thus, outside the cell separated A and B chains are essentially nontoxic, because the inherently toxic A chain lacks the ability to bind to cell surfaces and enter the cells in the absence of the B chain. Moreover, preproricin and proricin are also non-toxic, since the activity of the A chain is inhibited in these precursors. It is also known that in ricintype toxins the B chain binds to cell surfaces by virtue of galactose recognition sites, which react with glycoproteins or glycolipids exposed on the cell surface. It has been suggested that the toxicity of the ricin A chain might be exploited in antitumor therapy by replacing the indiscriminately binding B chain with a different targeting component having the ability to bind only to tumor cells. Thus, various immunotoxins have been prepared consisting of a conjugate of whole ricin or a separated ricin A chain and a tumorspecific monoclonal antibody or other targeting component while previously described immunotoxins comprising ricin are generally suitable for their specific purposes, they possess certain inherent limitations that detract from their overall utility in treating viral infections. One problem with the known conjugates arises from a structural feature of the A chain from natural ricin. It is known that the natural ricin A chain becomes Nglycosylated during its synthesis, by enzymes present in Ricinus cells, and it is thought that the resulting sugar moieties are capable of nonspecific interactions with cell surfaces. Thus, it appears that the known A chain conjugates are capable of a certain amount of binding with non target cells, even in the absence of the natural B chain, thus increasing the toxicity of such immunotoxins toward non-target cells. To partially mitigate this problem, recombinant A chain that lacks carbohydrate residues has been produced in E. coli. S. H. Pincus and V. V. Tolstikov, Anti-Human Immunodeficiency Virus Immunoconjugates, 32 Adv. Pharmacol. 205-247 (1995). Another problem with many ricin immunoconjugates arises from the fact that the B chain seems to have an important secondary function in that it somehow assists in the intoxication process, apart from its primary function in binding the ricin molecule to the cell surfaces. This secondary function is lost if the B chain is replaced by a different targeting component, such as a monoclonal antibody. Some researchers have addressed this problem by covalent attachment of affinity reagents to the B chain such that the galactose binding sites are blocked. J. M. Lambert et al., An Immunotoxin Prepared with Blocked Ricin: a Natural Plant Toxin Adapted for Therapeutic Use, 51 Cancer Res. 6236-6242 (1991).
The aforementioned modifications of ricin seek to enhance binding specificity to the outer cell surface by immunotoxins and similar, targeted therapeutic agents. Since certain types of infected cells do not express infection-related surface antigens, such binding specificity represents an inherent limitation. S. H. Pincus and V. V. Tolstikov, supra. A targeting-independent agent with a well-defined toxin activation mechanism involving a viral protease would permit the use of nonspecific xe2x80x9ctargetingxe2x80x9d (i.e., cell-binding) molecules, including sugar moieties and fully active ricin B chain. Therapeutic agents designed in this manner could eliminate a broader spectrum of infected cells, with potentially fewer undesirable side effects. Anti-HIV immunotoxins have been described that include antibodies linked to various toxic moieties via a peptide linker that includes a sequence cleavable by HIV protease. S. H. Pincus and V. V. Tolstikov, supra. In some cases, release of the toxic moiety by this protease may render it active, although the specific activation mechanism was not further defined. In the present invention, antibodies or segments thereof are only one of many potential targeting molecules for the therapeutic agents. Moreover, the activation mechanisms are clearly specified in the present invention. One such mechanism relies on protease-dependent cleavage at or near the natural protease activation site for a given toxin, not merely on release from a bulky xe2x80x9ccarrierxe2x80x9d protein (i.e., antibody). S. H. Pincus and V. V. Tolstikov, supra. In the case of ricin, the natural site for cleavage by proteolytic activity in Ricinus is in a disulfide-circumscribed loop in which one cysteine resides on the A chain and the other resides on the B chain; cleavage yields A and B chains connected by a disulfide bond. Therefore, most embodiments of the present invention that involve ricin include an HIV-protease cleavage sequence fused in-frame to the C-terminus of A chain such that the natural cleavage site is replaced with the HIV-protease site in the disulfide-circumscribed loop. In these embodiments, at lease some minimal N-terminal sequence of B chain required to inhibit A chain activity is retained, such that activation requires proteolytic cleavage and reduction of the disulfide bond. In all remaining embodiments, the mechanism of activation involves cleavage of a peptide linker to A chain, thereby separating adenine-like moieties that are chemically attached to the linker. Separation of the adenine-like residues unblocks the active site of ricin and allows A chain activity. Further, the foregoing text describes preferred embodiments (i.e., full B chain functionality, sugar moieties) that are highly compatible with these activation mechanisms. Indeed, these preferred embodiments are not suggested by others. S. H. Pincus and V. V. Tolstikov, supra. Certain embodiments of the present invention comprise attachment of hydrophobic moieties for intracellular targeting to sites of viral protease activity, which is limited in the cytosol. The aforementioned immunotoxins do not possess this aspect of the invention. While attachment of hydrophobic fatty acids to ricin A chain has been presented in terms of enhancing translocation across cell membranes for hypothetical medical applications, a method for activating ricin was not presented. A. V. Kabanov et al., Fatty Acylation of Proteins for Translocation Across Cell Membrane, 1 Biomed. Sci. 33-36 (1990); V. Y. Alakhov et al., Increasing Cytostatic Effects of Ricin A Chain and Staphylococcus aureus Enterotoxin A Through In Vitro Hydrophobization with Fatty Acid Residues, 12 Biotechnol. Appl. Biochem. 94-98 (1990). Hydrophobization of ricin is likely to increase toxicity to non-target cells, even if cell-surface targeting moieties are attached. A separate, viral-protease-dependent mechanism for activating ricin (and similar toxins) would prevent nonspecific toxicity. The present invention combines such a mechanism with hydrophobization.
In view of the foregoing, it will be appreciated that providing an antiviral agent that is activated only in cells infected with a selected virus, is non-toxic in uninfected cells, and is targeted independently of infection-related antigens, would be a significant advancement in the art. Furthermore, the prior art teaches away from making the present invention because specific embodiments described herein have previously been described as deleterious (B chain activity). Moreover, the prior art fails to describe or suggest elements of the present invention (e.g., means for fatty acid attachment) in combination with a protease-dependent toxin activation mechanism.
It is an object of the present invention to provide an antiviral agent that is toxic to virus-infected cells, but non-toxic to uninfected cells.
It is also an object of the invention to provide an antiviral agent for treating viral infections wherein the virus encodes a protease that is essential to virus replication.
It is another object of the invention to provide a toxin (e.g., ricin) based antiviral agent that maintains the inhibitory functionality of the lectin B chain, whereby the lectin B chain (or a portion thereof) inhibits the activity of the ricin A chain prior to proteolytic cleavage of a linker sequence and reduction of the disulfide bond. It is another object of the invention to provide a ricin- (or similar toxin) based antiviral agent that maintains the galactose-binding functionality of the ricin B chain, which enhances the binding of the antiviral agent to galactose residues on cell surfaces and the cellular internalization of the antiviral agent.
It is still another object of the invention to provide an antiviral agent for treating retroviral infections, including HIV infections.
It is yet another object of the invention to provide a method for treating retroviral infections wherein the virus encodes a protease that is essential to virus replication.
It is also an object of the invention to provide a method for treating HIV infections.
These and other objects can be addressed by providing a composition comprising a compound represented by the formula (Tm-AXB)-Hn or (A-X-B-Tm)Hn, wherein A is a protein synthesis inactivating toxin that is inactive until X is digested; X is a peptide susceptible to digestion by a viral protease; B is a lectin, T is a targeting moiety, H is a hydrophobic agent, m is 0 or an integer of at least 1, and n is 0 or an integer of at least 1. In a preferred embodiment of the invention, A is a ricin A chain and B is a ricin B chain or segment thereof. In another preferred embodiment of the invention, X is susceptible to digestion by a retroviral protease, such as a human immunodeficiency virus protease. In an especially preferred embodiment, X is a member selected from the group consisting of SEQ ID NO:12 and SEQ ID NO: 13. Preferably, the targeting moiety is a member selected from the group consisting of antigen-binding proteins, viral surface components and segments thereof, growth factors, lectins, and carbohydrates. Especially preferred targeting moieties include a member selected from the group consisting of antigen-binding proteins, viral surface components and segments thereof, proteins that bind viral surface components, growth factors, lectins, and carbohydrates. For example, such targeting moieties can include a member selected from the group consisting of antibodies against HIV glycoprotein gp120, antibodies against gp41, and the CD4 protein or segments thereof. As a further example, the targeting moiety can be an antigen-binding protein that binds the CD4 glycoprotein, such as gp120 or a segment thereof. Still another illustrative targeting moiety is a GAG protein segment.
The hydrophobic agent is preferably a member selected from the group consisting of bile acids, sterols, and saturated and unsaturated fatty acids. Preferred bile acids include cholic acid, deoxycholic acid, chenodeoxycholic acid, lithocholic acid, ursocholic acid, ursodeoxycholic acid, isoursodeoxycholic acid, lagodeoxycholic acid, glycocholic acid, taurocholic acid, glycodeoxycholic acid, glycochenodeoxycholic acid, dehydrocholic acid, hyocholic acid, hyodeoxycholic acid, and mixtures thereof. Preferred sterols include cholestanol, coprostanol, cholesterol, epicholesterol, ergosterol, ergocalciferol, and mixtures thereof. Preferred saturated or unsaturated fatty acid comprise about 4 to 20 carbon atoms, such as butyric acid, valeric acid, caproic acid, caprylic acid, capric acid, lauric acid, myristic acid, palmitic acid, stearic acid, oleic acid, linoleic acid, linolenic acid, eleostearic acid, and mixtures thereof.
Another preferred embodiment of the invention further comprises a pharmaceutically acceptable carrier admixed with the compound.
Still another preferred embodiment of the invention comprises a composition comprising a compound represented by the formula N-X-A or A-X-N, wherein A is a protein synthesis inactivating toxin that is inactive until digestion of X, X is a peptide susceptible to digestion by a viral protease, and N is an adenine moiety or a functional equivalent thereof. Yet another preferred embodiment of the invention comprises a method for treating a human immunodeficiency virus infection comprising administering an effective amount of a composition comprising:
(a) a member selected from the group consisting of:
(i) a compound represented by the formula (Tm-AXB)Hn or (AXBTm)Hn, wherein A is a protein synthesis inactivating toxin that is inactive until X is digested; X is a peptide susceptible to digestion by a human immunodeficiency virus protease; B is a lectin, T is a targeting moiety, H is a hydrophobic agent, m is 0 or an integer of at least 1, and n is 0 or an integer of at least 1,
(ii) a compound represented by the formula N-X-A or A-X-N, wherein A is a protein synthesis inactivating toxin that is inactive until digestion of X, X is a peptide susceptible to digestion by a human immunodeficiency virus protease, and N is an adenine moiety or functional equivalent thereof, and
(iii) mixtures of (i) and (ii); and
(b) a pharmaceutically acceptable carrier.
A further preferred embodiment of the invention comprises a nucleic acid encoding a peptide represented by the formula A-X-B wherein A is a protein synthesis inactivating toxin that is inactive until digestion of X; X is a peptide susceptible to digestion by a viral protease; and B is a lectin or other targeting moiety.
Before the present antiviral agents and methods of use thereof for treating viral infections are disclosed and described, it is to be understood that this invention is not limited to the particular configurations, process steps, and materials disclosed herein as such configurations, process steps, and materials may vary somewhat. It is also to be understood that the terminology employed herein is used for the purpose of describing particular embodiments only and is not intended to be limiting since the scope of the present invention will be limited only by the appended claims and equivalents thereof.
The publications and other reference materials referred to herein to describe the background of the invention and to provide additional detail regarding its practice are hereby incorporated by reference. The references discussed herein are provided solely for their disclosure prior to the filing date of the present application. Nothing herein is to be construed as an admission that the inventors are not entitled to antedate such disclosure by virtue of prior invention.
It must be noted that, as used in this specification and the appended claims, the singular forms xe2x80x9ca,xe2x80x9d xe2x80x9can,xe2x80x9d and xe2x80x9cthexe2x80x9d include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to a composition comprising xe2x80x9ca carrierxe2x80x9d includes reference to two or more of such carriers, reference to xe2x80x9ca fatty acidxe2x80x9d includes reference to one or more of such fatty acids, and reference to xe2x80x9ca targeting moietyxe2x80x9d includes reference to two or more of such targeting moieties.
In describing and claiming the present invention, the following terminology will be used in accordance with the definitions set out below as used herein, xe2x80x9ccomprising,xe2x80x9d xe2x80x9cincluding,xe2x80x9d xe2x80x9ccontaining,xe2x80x9d xe2x80x9ccharacterized by,xe2x80x9d and grammatical equivalents thereof are inclusive or open-ended terms that do not exclude additional, unrecited elements or method steps. xe2x80x9cComprisingxe2x80x9d is to be interpreted as including the more restrictive terms xe2x80x9cconsisting ofxe2x80x9d and xe2x80x9cconsisting essentially of.xe2x80x9d
As used herein, xe2x80x9cconsisting ofxe2x80x9d and grammatical equivalents thereof exclude any element, step, or ingredient not specified in the claim.
As used herein, xe2x80x9cconsisting essentially ofxe2x80x9d and grammatical equivalents thereof limit the scope of a claim to the specified materials or steps and those that do not materially affect the basic and novel characteristic or characteristics of the claimed invention.
As used herein, xe2x80x9cprotein synthesis inactivating toxinxe2x80x9d includes toxins that are ribonucleases, N-glycosidases, or ADP-ribosyltransferases. N-glycosidases are exemplified by the single polypeptide of the plant type I ribosome inactivating proteins (e.g., gelonin, momordin, and saporin), and the xe2x80x9cAxe2x80x9d chain of the plant type II ribosome-inactivating proteins (e.g., ricin, abrin, modeccin, and the like) and similar acting bacterial toxins. xe2x80x9cProtein synthesis inactivating toxinxe2x80x9d also includes the specific ribonucleases that digest a specific phosphodiester bond in the backbone of ribosomal RNA, thereby inactivating the ribosomes and inhibiting protein synthesis. Ribonucleases are exemplified by the fungal toxins alpha-sarcin, mitogillin, and restrictocin, but also include similar acting bacterial toxins. xe2x80x9cProtein synthesis inactivating toxinxe2x80x9d also includes the ADP-ribosylating component of the ADP-ribosyltransferases, which are proteolytically activated bacterial toxins that ADP-ribosylate, and thus inactivate, components of the protein synthesis machinery (e.g., diphtheria toxin, Pseudomonas exotoxin A).Plant ribosome-inactivating proteins (RIPs) are N-glycosidases that cleave the N-glycosidic bond of adenine in a specific ribosomal RNA sequence. Many RIPs are single-chain proteins (type I RIPs), but some (type II RIPs) possess a galactose-specific lectin domain that binds to cell surfaces. J. M. Ford, M. R. Hartley, L. M. Roberts, Ribosome Inactivating Proteins of Plants, 2 Seminars in Cell Biology 15-22 (1991). The type II RIPs are potent toxins, the best known of which is ricin. As used herein, xe2x80x9ctype II ribosome-inactivating proteinsxe2x80x9d or xe2x80x9ctype II RIPsxe2x80x9d means two-chain N-glycosidases that cleave the N-glycosidic bond of adenine in a specific ribosomal RNA sequence, wherein the two chains are an A chain, which possesses the N-glycosidase activity, and a B chain, which comprises a galactose-specific lectin domain that binds to cell surfaces. Ricin is the prototypical type II ribosome-inactivating protein, but other such type II RIPs include abrin (from Abrus precatrius), modeccin (from Adenia digtata), viscumin (from Viscum album), and volkensin (from Adenia volkensii). C. H. Hung et al., Cloning and Expression of Three Abrin A-chains and Their Mutants Derived by Site-specific Mutagenesis in Escherichia coli, 219 Eur. J. Biochem. 83-87 (1994); K. A. Wood et al., Preproabrin: Genomic Cloning, Characterisation and the Expression of the A-chain in Escherichia coli, 198 Eur. J. Biochem. 723-732 (1991); C. H. Hung et al., Primary Structure of Three Distinct Isoabrins Determined by cDNA Sequencing: Conservation and Significance, 229 J. Mol. Biol. 263-267 (1993); GenBank accession numbers X76644, X76720, X76721, X76722, X54873, X54872, X55667, A58957; International Application publication no. W09701636.As used herein, xe2x80x9cricin A chainxe2x80x9d means an N-glycosidase of about 32 kDa that digests and inactivates 26S and 28S ribosomal RNA by cleavage of a specific adenine residue located within a highly conserved region of the 26S and 28S ribosomal RNA, as is well known in the art. As used herein, xe2x80x9cricin B chainxe2x80x9d means a galactose/N-acetylgalactosamine-binding lectin of about 34 kDa. SEQ ID NO:1 shows a DNA encoding preproricin and translation product thereof, wherein the signal peptide comprises amino acid residues xe2x88x9224 to xe2x88x921, the A chain comprises amino acid residues 1 to 267, the linker peptide comprises amino acid residues 268 to 279, and the B chain comprises amino acid residues 280 to 541. This sequence is incomplete at the 5xe2x80x2 end (L. M. Roberts, J. W. Tregear, J. M. Lord, Molecular Cloning of Ricin, 7 Targeted Diagn Ther 81-97 (1992)) and the entire sequence is given in SEQ NO. 2 and SEQ NO. 3, which follows. SEQ ID NO:2 and SEQ ID NO:3 show other DNAs encoding preproricin and their translation products thereof wherein the signal peptides comprise amino acid residues 35 to xe2x88x921, the A chains comprise amino acid residues 1 to 267, the linker peptides comprise amino acid residues 268 to 279, and the B chains comprise amino acid residues 280 to 541. Preferred ricin A chains and ricin B chains include the peptides having the amino acid sequences identified as SEQ ID NO:1, SEQ ID NO:2, SEQ ID NO:3, and biologically functional equivalents thereof. These functional equivalents include any two-subunit, proteolytically activatable, N-glycosidase toxins that inactivate ribosomes, i.e., type II ribosome-inactivating proteins. Such functional equivalents of the ricin A chain, for example, retain functionality in digesting ribosomal RNA although they may be truncations, deletion variants, or substitution variants of the ricin A chain portions of SEQ ID NO:1, SEQ ID NO:2, or SEQ ID NO:3 or include additional amino acid residues attached thereto. Functional equivalents of the ricin A chain may also be derived from any other type I or type II ribosome inactivating protein. Functional equivalents of the ricin B chain, for example, (a) retain functionality of binding to galactose residues on the surface of cells although they may be truncations, deletion variants, or substitution variants of the ricin B chain portions of SEQ ID NO:1, SEQ ID NO:2, or SEQ ID NO:3 or include additional amino acid residues attached thereto; (b) retain functionality for inhibiting activity of the ricin A chain; (c) or both. Carbohydrate moieties may optionally be attached to the A or B chains.
As mentioned above, changes may be made in the structure of the type II ribosome-inactivating protein A chain or B chain while maintaining the desirable N-glycosidase or galactose-binding characteristics thereof. For example, certain amino acid residues may be substituted for other amino acid residues in a protein structure without appreciable loss of activity. Since it is the interactive capacity and nature of a protein that defines that protein""s biological functional activity, certain amino acid sequence substitutions can be made in a peptide""s sequence and nevertheless obtain a protein with like properties. It is thus contemplated that various changes may be made in the sequence of a type II ribosome-inactivating protein A chain or B chain without appreciable loss of its biological utility or activity.
It is also well understood by the skilled artisan that inherent in the definition of a biologically functional equivalent protein or peptide is the concept that there is a limit to the number of changes that may be made within a defined portion of the molecule and still result in a molecule with an acceptable level of equivalent biological activity. It is also well understood that where certain residues are shown to be particularly important to the biological or structural properties of a protein or peptide, e.g. residues in active sites, such residues may not generally be exchanged. Amino acid substitutions are generally based on the relative similarity of the amino acid side-chains relative to, for example, their hydrophobicity, hydrophilicity, charge, size, and the like. An analysis of the size, shape, and type of the amino acid side-chains reveals, for example, that arginine, lysine, and histidine are all positively charged residues at neutral pH; that alanine, glycine, and serine are all similar in size; and that phenylalanine, tryptophan, and tyrosine all have a generally similar shape. Therefore, based upon these considerations, the following conservative substitution groups or biologically functional equivalents have been defined:(a) Cys; (b) Phe, Trp, Tyr; (c) Gln, Glu, Asn, Asp; (d) His, Lys, Arg; (e) Ala, Gly, Pro, Ser, Thr; and (f) Met, Ile, Leu, Val. M. Dayhoff et al., Atlas of Protein Sequence and Structure (Nat""l Biomed. Res. Found., Washington, D.C., 1978). Based on similar considerations, another grouping of amino acids is as follows: (a) Pro; (b) Ala, Gly; (c) Ser, Thr; (d) Asn, Gln; (e) Asp, Glu; (f) His; (g) Lys, Arg; (h) Cys; (i) Ile, Leu, Met, Val; and (j) Phe, Trp, Tyr. M. Jimenez-Montano and L. Zamora-Cortina, Evolutionary Model for the Generation of Amino Acid Sequences and its Application to the Study of Mammal Alpha-hemoglobin Chains, Proc. VIIth Int""l Biophysics Congress, Mexico City (1981).
To effect more quantitative changes, the hydropathic index of amino acids may be considered. Each amino acid has been assigned a hydropathic index on the basis of its hydrophobicity and charge characteristics, which are as follows: isoleucine (+4.5); valine (+4.2); leucine (+3.8); phenylalanine (+2.8); cysteine (+2.5); methionine (+1.9); alanine (+1.8); glycine (0.4); threonine (0.7); serine (0.8); tryptophan (0.9); tyrosine (1.3); proline (1.6); histidine (3.2); glutamate (3.5); glutamine (3.5); aspartate (3.5); asparagine (3.5); lysine (3.9); and arginine (4.5).The importance of the hydropathic amino acid index in conferring interactive biological function on a protein is generally understood in the art. J. Kyte and R. Dooliffle, A Simple Method for Displaying the Hydropathic Character of a Protein, 157 J. Mol. Biol. 105-132 (1982). It is known that certain amino acids may be substituted for other amino acids having a similar hydropathic index or score and still retain a similar biological activity. In making changes based on the hydropathic index, the substitution of amino acids whose hydropathic indices are within xc2x12 is preferred, within xc2x11 is particularly preferred, and within xc2x10.5 is even more particularly preferred.
It is also understood that an amino acid can be substituted for another having a similar hydrophilicity value and still obtain a biologically equivalent protein. As detailed in U.S. Pat. No. 4,554,101, the following hydrophilicity values have been assigned to amino acid residues: arginine (+3.0); lysine (+3.0); aspartate (+3.0); glutamate (+3.0xc2x11); serine (+0.3); asparagine (+0.2); glutamine (+0.2); glycine (0); threonine (0.4); proline (0.5xc2x11); alanine (0.5); histidine (0.5); cysteine (1.0); methionine (1.3); valine (1.5); leucine (1.8); isoleucine (1.8); tyrosine (2.3); phenylalanine (2.5); tryptophan (3.4).
In making changes based upon similar hydrophilicity values, the substitution of amino acids whose hydrophilicity values are withinxc2x12 is preferred, within xc2x11 is particularly preferred, and withinxc2x10.5 is even more particularly preferred.
As used herein, xe2x80x9cviral proteasexe2x80x9d and similar terms mean proteases coded for by viral nucleic acids and expressed in virus-infected cells. For example, all known human retroviruses encode their own proteases. E.g., R. A. Katz and A. M. Skalka, The Retroviral Enzymes, 63 Annu. Rev. Biochem. 133-173 (1994); K. Von der Helm, Retroviral Proteases: Structure, Function and Inhibition from a Non-anticipated Viral Enzyme to the Target of a Most Promising HIV Therapy, 377 Biol. Chem. 765-774 (1996). Retroviral proteases cleave a polyprotein precursor into a reverse transcriptase, capsid proteins, and other processed products. A variety of other viruses also encode their own proteases, including herpesviruses, hepatitis C viruses, rhinoviruses, and picomaviruses. A. K. Patick and K. E. Potts, Protease Inhibitors as Antiviral Agents, 11 Clin. Microbiol. Rev. 614-627 (1998); B. D. Korant, Viral Proteases: An Emerging Therapeutic Target, 8 Crit. Rev. Biotechnol. 149-157 (1988).
As used herein, xe2x80x9chuman immunodeficiency virus protease cleavage sitexe2x80x9d and similar terms mean peptides that are digested by HIV protease. HIV-1 and HIV-2 proteases, although not identical, both cleave a viral GAG precursor protein of HIV-2 at two very different sites to yield the same products. J. C. Wu et al., Synthetic HIV-2 Protease Cleaves the GAG Precursor of HIV-1 with the Same Specificity as HIV-1 Protease, 277 Arch. Biochem. Biophys. 306-311 (1990). One such site in the GAG precursor is Val-Ser-Gln-Asn-Tyr-Pro-Ile-Val-Gln-Asn (type 1 cleavage site wherein cleavage occurs between the tyrosine and proline residues; SEQ ID NO:4). The second site is Ser-Ala-Thr-Ile-Met-Met-Gln-Arg-Gly-Asn (type 2 cleavage site wherein cleavage occurs between the two methionine residues, SEQ ID NO:11). There are many different sequences cleaved by HIV-1 protease, which generally fall into one of two types exemplified by these two sites: type 1 having Tyr(Phe)-Pro- and type 2 having hydrophobic residues (excluding proline) at the site of cleavage. J. Txc3x6zsxc3xa9r et al., Studies on the Symmetry and Sequence Context Dependence of the HIV-1 Proteinase Specificity, 272 J. Biol. Chem. 16807-16814 (1997). This article by Txc3x6zsxc3xa9r et al. goes on to conclude that classification of retroviral cleavage sites into two types is an oversimplification and the strong sequence context dependence also raises difficulties for predicting cleavage sites. While the variability of sequences cleaved by HIV protease makes it difficult to predict cleavage sites, models have been successfully generated for the purpose of predicting these sites. K. C. Chou, A Vectorized Sequence-coupling Model for Predicting HIV Protease Cleavage Sites in Proteins, 268 J. Biol. Chem. 16938-16948 (1993); K. C. Chou, Prediction of Human Immunodeficiency Virus Protease Cleavage Sites in Proteins, 233 Anal. Biochem. 1-14 (1996). Proteases from HIV-1 and HIV-2 recognize and cleave the same two types of sequences, although not necessarily the same sequences with the same cleavage efficiencies. Thus, it is possible to construct a therapeutic agent useful for both forms of the virus. HIV-1 is the form prevalent in the Western world, while HIV-2 is typically found in West African patients with AIDS. However, HIV-1 is likely to cause AIDS in West African patients as well.
As used herein, xe2x80x9cpeptidexe2x80x9d means peptides of any length and includes proteins. The terms xe2x80x9cpolypeptidexe2x80x9d and xe2x80x9coligopeptidexe2x80x9d are used herein without any particular intended size limitation, unless a particular size is otherwise stated.
As used herein, xe2x80x9ccarbohydratexe2x80x9d means carbohydrate monomers, oligomers, and polymers. There is no particular intended size limitation with respect to carbohydrate oligomers or polymers unless a particular size is otherwise stated.
As used herein, xe2x80x9clectinxe2x80x9d means a class of proteins of nonimmunological origin that binds carbohydrates. The scope of lectins that can be used according to the present invention is limited only by functionality, i.e., binding to carbohydrates and/or inhibiting activity of a protein synthesis inactivating toxin to which it is fused. Segments or portions of lectins are also to be considered within the scope of the term xe2x80x9clectinxe2x80x9d provided that such segments or portions retain the carbohydrate-binding and/or toxin-inhibiting function. Representative lectins that can be used according to the present invention include the following: abrin or jequirity bean (Abrus precatorius), asparagus pea or lotus or winged pea (Tetragonolobus purpureas), avocado (Persea americana), bitter pear melon (Momordica charantia), broad bean (Vicia faba), camels foot tree (Bauhinia purpurea), castor bean (Ricinus communis), chick pea (Cicer arietinum), Mozambique cobra (Naja mocambique mocambique), Thailand cobra (Naja naja laculbia), concanavalin A or jack bean (Canavalia ersiformis), Israel coral tree (Erythina oxaliodendron), daffodil (Narcissus pseudonarcissus), eel (Anguilla anguilla), elderberry (Sambucus nigra), furze or gorse (Ulex europaeus), green marine algae (Cocleum fragile), hairy vetch (Vicia villosa), horse gram (Dolichos biflorus), horseshoe crab or limulin (Limulus polyphemus), jacalin (Artocarpus integrifolia), Japanese wisteria (Wisteria floribunda), jimson weed or thorn apple (Datura stramonium), Scotch laburnum (Laburnum alpinum), lentil (Lens culinaris), lima bean (Phaseolus limensis), European mistletoe or viscumin (Viscum album), mung bean (Vigna radiata), mushroom (Agaricus bisporus), Osage orange (Maclura pomifera), pagoda tree (Sophora japonica), pea (Pisum sativum), peanut (Arachis hypogaea), pokeweed (Phytolacca americana), potato (Solanum tuberosum), red kidney bean (Phaseolus vulgaris), red marine algae (Plifola plumosa), Roman snail (Helix porrata), scarlet runner bean (Phaseolus coccineus), Scotch broom (Cytisus scoparius), Siberian pea tree (Caragana arborescens), snail-edible (Helix pomatia), snail-garden (Helix aspersa), snowdrop (Galanthus nivalis), soybean (Glycine max), spindle tree (Euonymus europaeus), sweet pea (Lathyrus odoratus), tomato (Lycopersicon esculentum), wheat germ (Triticum vulgaris), winged bean (Psophoramus beliagonolobus). Preferred lectins include the B chains of type II RIPS, such as ricin, abrin, modeccin, viscumin, and volkensin. Additional preferred lectins include tora-mame lectin (from Phaeolus vulgaris), phytohemagglutinin (PHA), wheat germ agglutinin, achatinin H, and Vicia villosa lectin. Still additional preferred lectins include lectins that bind to T lymphocytes, including mistletoe lectin, Phaseolus vulgaris leucoagglutinin, and lectins from Bandeiraea simplicifolia BS-I, Bauhinia purpurea, Glycine max, Lycopersicon esculentum, Triticum vulgaris, Canavalia ensiformis, Lens culinaris, Phaseolus coccineus, and Wisteria floribunda. Further additional preferred lectins include lectins that bind to sulfated carbohydrates, such as L-selectin and human galectin 1. Still further preferred lectins include lectins that bind to HIV gp120, including concanavalin A, wheat germ agglutinin, Lens culinaris agglutinin, Vicia faba agglutinin, Pisum sativum agglutinin, and phytohaemagglutinin. Another group of preferred lectins include the B subunits of Shiga and Shiga-like toxins (SLTs), such as the B chains of Shiga toxin, SLT-I, SLT-HI, SLT-IIv, and SLT-2e. Still other preferred lectins include jacalin, mannose binding proteins (e.g., concanavalin A), and maltose binding protein.
As used herein, a xe2x80x9cpharmaceutically acceptablexe2x80x9d component is one that is suitable for use with humans and/or animals without undue adverse side effects (such as toxicity, irritation, and allergic response) commensurate with a reasonable benefit/risk ratio. Such materials are pharmaceutically acceptable in that they are nontoxic, do not interfere with drug delivery, and are not for any other reasons biologically or otherwise undesirable.
As used herein, xe2x80x9ceffective amountxe2x80x9d means an amount of a drug or pharmacologically active agent that is nontoxic but sufficient to provide the desired local or systemic effect and performance at a reasonable benefit/risk ratio attending any medical treatment.
As used herein, xe2x80x9cadministeringxe2x80x9d and similar terms mean delivering the composition to the individual being treated such that the composition is capable of being circulated systemically to the parts of the body where the viral protease-sensitive linker can be digested by the viral protease, thus cleaving a peptide bond in a linker that connects the type II ribosome-inactivating protein A chain and the type II ribosome-inactivating protein B chain. After such digestion, however, the A chain and the B chain will still be coupled to each other by a disulfide bond. Thus, the composition is preferably administered to the individual by systemic administration, typically by subcutaneous, intramuscular, or intravenous administration, or intraperitoneal administration. Injectables for such use can be prepared in conventional forms, either as a liquid solution or suspension or in a solid form suitable for preparation as a solution or suspension in a liquid prior to injection, or as an emulsion. Suitable excipients or carriers include, for example, water, saline, dextrose, glycerol, ethanol, and the like; and if desired, minor amounts of auxiliary substances such as wetting or emulsifying agents, buffers, and the like can be added. Other carriers can be used and are well known in the art.
In its most basic embodiment, the present invention comprises a composition and method of use thereof, wherein the composition comprises a type II ribosome-inactivating protein precursor protein, such as proricin, wherein the natural linker is replaced with a viral-protease-cleavable linker. In an especially preferred embodiment of the invention, the natural linker is replaced with an HIV-protease-cleavable linker. Generally, there are two types of sequences cleaved by HIV proteases, which can be identified with some predictability. In certain embodiments, a major advantage of, for example, a proricin variant over typical immunotoxins is that cells in early stages of infection can be killed, thus preventing HIV replication.
Targeting molecules can be added to various sites on the proricin-based agent in addition to or for replacing the targeting activity of the B chain. Targeting molecules include, but are not limited to, CD4 and derivatives thereof; antibodies such as anti-gp120, anti-gp41, and the like; IL-2 segments; gp120 segments; gag segments; fatty acid or other hydrophobic substituents; and mannose-containing carbohydrate moieties. Especially preferred targeting molecules include gp120 segments and fatty acid or other hydrophobic substituents. Fatty acid groups may permit direct translocation of the agent into cells, rendering the activity of the B chain unnecessary, even though the presence of the B chain or a segment thereof would remain important so that the A chain would be inactive until proteolytically activated by HIV protease. An alternative mechanism for proteolytic activation involves an adenine or similar group attached to an HIV-protease-cleavable extension linker on the A chain, which will inhibit A chain activity until cleaved by HIV protease.
A major advantage of the present invention is that the antiviral agent is activated in viral particles or early-stage infected cells, killing the cells upon infection and effectively preventing the integration of the viral genome into the host genome and preventing the latency/rebound problem. Another major advantage in certain embodiments is that the present invention should be able to enter all HIV susceptible cells, not just cells known to act as host cells for the virus. Moreover, the agent remains inert in a cell until degraded therein, unless the cell is infected with the virus, wherein the viral protease activates the agent.
Human immunodeficiency virus (HIV) is representative of viruses that encode their own protease enzymes that cleave specific sites on viral proteins during the replication process. J. Gatlin et al., Regulation of Intracellular Human Immunodeficiency Virus Type-1 Protease Activity, 244 Virology 87-96 (1998). Examples of other protease-encoding viruses are yellow fever virus and tick-borne encephalitis virus. T. J. Chambers et al., Evidence that the N-terminal Domain of Nonstructural Protein NS3 from Yellow is Fever Virus is a Serine Protease Responsible for Site-specific Cleavages in the Viral Polyprotein, 87 Proc. Nat""l Acad. Sci. USA 8898-8902 (1990); K. V. Pugachev et al., Site-directed Mutagenesis of the Tick-borne Encephalitis Virus NS3 Gene Reveals the Putative Serine Protease Domain of the NS3 Protein, 328 FEBS Lett. 115-118 (1993). Vaccinia virus, which is so closely related to the smallpox virus that it was used as the basis for a vaccine against smallpox, apparently uses a protease specifically located within infected cells, R. M. L. Buller et al., Poxvirus Pathogenesis, 55 Microbiol. Rev. 80-122 (1991). The smallpox virus may use a similar protease.
The therapeutic agent described herein is based on type II ribosome-inactivating proteins, such as the protein toxin, ricin, and is generally applicable to viruses that encode site-specific proteases. From this point in the application, ricin will be used as a preferred example of a type II ribosome-inactivating protein, but it is intended that the scope of the invention include any protein synthesis inactivating toxin, as previously defined. As briefly reviewed above, ricin from castor beans comprises two subunits (A and B) connected by a disulfide bond, with carbohydrate chain(s) attached to each. J. M. Lord et al., Ricin: Structure, Mode of Action, and Some Current Applications, 8 FASEB J. 201-208 (1994); U.S. Pat. No. 5,622,838. The A subunit is an enzyme that cleaves a critical adenine residue from ribosomal RNA, thereby inactivating ribosomes. The B subunit is a lectin that binds to cell-surface galactose residues and also contributes to cellular internalization of the A-B toxin.
The rationale for using ricin-based therapeutic agents against HIV has been documented. M. G. C. T. Van Oijen et al., Rationale for Use of Immunotoxins in the Treatment of HIV-infected Humans, 5 J. Drug Target 75-91 (1997). Viruses appropriate the metabolic machinery of infected host cells to replicate themselves. By inactivating the machinery for protein synthesis, i.e., ribosomes, and thereby killing infected cells, selective ricin-based agents can prevent the replication of HIV. This strategy relies on is attaching targeting molecules that bind to infected cells. There are two major drawbacks to this approach. First, the viral antigens to which the targeting molecules bind may either be shed from the cell surfaces, J. M. Lord et al., supra, or occur on free viral particles, not just cell surfaces. Thus the intended targets (infected cells) are not efficiently destroyed. Second, the ricin-plus-targeting molecule conjugates must be administered in large doses, presumably because they are not ideally suited to the complex internalization pathway whereby ricin enters cells. B. M. Simmons et al., Mannose Receptor-mediated Uptake of Ricin Toxin and Ricin A Chain by Macrophages, 261 J. Biol. Chem. 7912-7920 (1986).
The present approach involves a modification of native proricin, which is the precursor of ricin that differs only in that the A and B subunits are connected by a peptide linker to form a continuous proricin peptide. Cleavage of proricin by a protease in castor beans (the source of ricin) yields the mature ricin protein. In the present invention, the naturally occurring peptide linker between the ricin A chain and the ricin B chain is replaced by an HIV-protease-cleavable site. Research results suggest that the A and B subunits are not normally separated until ricin reaches the cytoplasm, wherein the connecting disulfide bond is cleaved, J. P. Frenoy et al., Uptake of Injected 125I-ricin by Rat Liver In Vivo, 284 Biochem. J. 249-257 (1992); hence, a modified peptide linker connecting the A and B subunits should not interfere with cellular internalization of the protein variant. Indeed, internalization of this variant should be significantly more efficient than for immunotoxins that include targeting molecules.
These data strongly suggest that incorporating an HIV-protease-sensitive linker into proricin will produce an agent that selectively kills IRV-infected cells. Moreover, the proricin variant should work against both forms of the virus, HIV-1 and HIV-2, because it has been shown that HIV-2 protease cleaves the polyprotein precursor of HIV-1 with the same specificity as the HIV-1 protease. J. C. Wu et al., supra.
The proricin variant will bind to any cell that displays galactose residues on the cell surface. Another route for cellular uptake involves the mannose-containing carbohydrate chains attached to ricin as it is produced in castor beans and as it should be produced in a yeast cell host. These mannose groups are bound by mannose receptors on cell surfaces. B. M. Simmons et al., supra. Additional routes for cellular uptake could be engineered by attaching other sugar-binding lectins to the N-terminus of the ricin A chain or the C-terminus of the B chain. Other proteins have been attached to the N-terminus of the A chain without affecting ricin activity or internalization. B. Beaumelle et al., Ricin A Chain Can Transport Unfolded Dihydrofolate Reductase into the Cytosol, 272J. Biol. Chem. 22097-22102 (1997). Together, the many possible proricin variants will be able to enter virtually any cell type, but will be activated within and kill only virus-infected cells. Although the preferred embodiment of the present invention, described above, is designed to be activated in cells infected with HIV, it is also possible that other proteases present in the healthy human body might activate the proricin-based therapeutic agent. Therefore, it would be advantageous to increase the selectivity of the therapeutic agent still further. One solution for increasing selectivity would be to insert a linker containing a type 2 HIV-protease-cleavable site into proricin instead of a type 1 cleavable site. A protease found naturally in the human body may cleave one of the linker types, and activate the therapeutic agent. By changing the linker sequence, selectivity for HIV protease may be increased by avoiding linker cleavage by naturally occurring proteases.
A second solution to avoiding activation by nonspecific proteases is to insert amino acid residues into natural, exposed loops of A chain that are also found in the cleavable linker; such insertions cannot significantly reduce A chain activity. Also, these insertions cannot contain sequences cleavable by the viral protease. Therefore, if a host (i.e., human) protease can cleave the linker at a sequence also found in an A-chain loop insertion, the host protease will inactivate the ricin A chain and prevent destruction of uninfected cells.
Another solution to increasing selectivity is to attach a targeting moiety to the proricin variant. Five categories of targeting molecules are discussed herein: (1) Targeting to the surfaces of infected cells displaying viral proteins as the targets, e.g., gp 120. Such targeting molecules would include antibodies, viral proteins or portions thereof, and other proteins that bind to such viral components (e.g., CD4). Typical immunotoxins use this type of targeting molecule. (2) Targeting to injectable or susceptible cells. Molecules used for targeting include proteins that bind to CD4. (3) Targeting to virtually any cell. The ricin B chain binds to any cell displaying galactose residues. Mannose chains on ricin A and B chains could bind to mannose receptors. Fatty acids permit internalization into a variety of cells. Lectins other than the ricin B chain can also be used. (4) Intracellular targeting to the inner surface of the cell membrane for incorporation into viral particles. Fatty acids and/or segments of the GAG protein attached to the antiviral agent will promote this process. Any protein segment to which fatty acids are attached by enzymes in vivo can be fused to the antiviral agent. (5) Targeting with viral vectors, such as an HIV-based vector that would ultimately insert DNA encoding the antiviral agent into susceptible cells. One or more of these five different types of targeting strategies can be used.
A preferred targeting molecule for attachment to the modified proricin of the present invention is a molecule that targets the CD4 glycoprotein on cell surfaces or an antibody against an HIV envelope glycoprotein (S. H. Pincus, Therapeutic Potential of Anti-HIV Immunotoxins, 33 Antiviral Res. 1-9 (1996)). As is well-known in the art, attachment of HIV to cells involves the interaction of the HIV gp120 envelope glycoproteins with specific receptors on cell surfaces-the CD4 glycoprotein and members of the chemokine receptor family. R. Wyatt and J. Sodroski, The HIV-1 Envelope Glycoproteins: Fusogens, Antigens, and Immunogens, 280 Science 1884-1888 (1998). The CD4 glycoprotein is expressed on the surface of T lymphocytes, monocytes, dendritic cells, and brain microglia, the main target cells for primate immunodeficiency viruses in vivo. Use of CD4 protein or a derivative thereof as a targeting molecule would cause the present invention to bind surface-exposed pg120 and enter HIV-infected cells and cause inactivation of ribosomes in such cells, thus destroying the ability of the cells to produce new HIV proteins. However, this approach would only target cells in the late stage of infection that are expressing viral proteins on the cell surface. The problem at that point is that the viral genome has already been incorporated into the host cell genome. If viral particles are not being actively produced and their proteins displayed on the cell surface, then the cell will avoid the toxic effects of the CD4-based proricin variant. The virus could rebound at a later date upon cellular activation. G. Mathe, The Kinetics of Cancer Cells and of HIV1: The Problems of Cell and Virus Rebounds and of Latency, 52 Biomed. Pharmacother. 413-420 (1998).
HIV infections of central nervous system (CNS) cells can involve galactosyl ceramide as a receptor, enabling virus entry by a CD4-independent pathway. S. H. Pincus and V. V. Tolstikov, supra. Furthermore, infected CNS cells express little or no viral antigens on their surfaces. While these are not major sites of viral replication, such infected cells may represent a barrier to complete elimination of viral replication. Classical immunotoxins are poor candidates for attacking such cells. In contrast, the present embodiments of the invention involving fully functional B chain could readily bind to galactosyl ceramide, yet remain inactive unless HIV protease is encountered. Further, a compatible embodiment involving fatty acid attachment would facilitate crossing of the blood-brain barrier, which is a major impediment to the efficacy of immunotoxins.
Another targeting molecule would be a portion of interleukin-2 (IL-2) that binds to the high-affinity IL-2 receptors found on a subset of activated T-lymphocytes thought to be the main or sole site of HIV replication. However, it has not been demonstrated that this is the sole site in infected humans. S. H. Pincus, supra. Furthermore, the expression of IL-2 receptors on the surface of T cells appears to result from HIV infection or the binding of HIV gp120 to CD4 on the T-cell surface, R. W. Finberg et al., Selective elimination of HIV-1-infected Cells with an Interleukin-2 Receptor-specific Cytotoxin, 252 Science 1703-1705 (1991). Therefore, use of a segment of IL-2 on a classical immunotoxin would target infected cells, although T cells activated by non-HIV stimulants would also be eliminated. S. H. Pincus and V. V. Tolstikov, supra. However, infected cells targeted in this manner should be at an earlier stage of infection relative to cells targeted by immunotoxins containing CD4 segments or antibodies against viral proteins. A portion of IL-2 has been fused to diphtheria toxin as an anti-HIV xe2x80x9cimmunotoxin.xe2x80x9d R. W. Finberg et al., supra.
Since use of IL-2 as a targeting molecule on classical immunotoxins could result in destruction of uninfected cells and deleterious health effects, the protease activation mechanism of the present invention will be important to confer additional selectivity. A strategy for attacking viral latency (stable, dormant provirus integrated into host DNA) is to activate infected T cells to express IL-2 receptors and perhaps viral proteins by exposing patients to exogenous HIV gp120. H. Kornfeld, W. W. Cruikshank, S. W. Pyle, J. S. Berman. D. M. Center, Lymphocyte Activation by HIV-1 Envelope Glycoprotein, 335 Nature 6189 (1988). After several hours, treatment with a protease-activated therapeutic agent containing IL-2 would destroy infected cells. Without activation of the provirus, the therapeutic agent would likely be degraded in the cell before it could be activated. Alternatively, use of the carbohydrate-(galactose-) binding B chain or another lectin separately or as part of the toxin-based therapeutic agent could activate infected T cells. B. A. Sela et al., Lymphocyte Activation by Monovalent Fragments of Antibodies Reactive with Cell Surface Carbohydrates, 143 J. Exp. Med. 665-671 (1976); O. Closs et al., Stimulation of Human Lymphocytes by Galactose-specific Abrus and Ricinus Lectins, 115 J. Immunol. 1045-1048 (1975).
A better approach would be to kill infected cells at a very early stage of infection, i.e. before the integration of the viral genome. To do this would require a different targeting molecule. One possibility is attaching an octapeptide (or derivative thereof) from the HIV envelope glycoprotein gp120. C. B. Pert et al., Octapeptides Deduced from the Neuropeptide Receptor-like Pattern of Antigen T4 in Brain Potently Inhibit Human Immunodeficiency Virus Receptor Binding and T-cell Infectivity, 83 Proc. Nat""l Acad. Sci. USA 9254-9258 (1986). This short sequence is believed to play an important role in HIV attachment to cells. It could be incorporated into the proricin variant at its N-terminus or at another site in the variant. The receptor for this octapeptide is the T4 or CD4 protein. Entry of primate immunodeficiency viruses into the host cell involves the binding of the gp120 envelope glycoprotein to the CD4 glycoprotein, which serves as the primary receptor. P. D. Kwong et al., Structure of an HIV gp120 Envelope Glycoprotein in Complex with the CD4 Receptor and a Neutralizing Human Antibody, 393 Nature 648-659 (1998). This receptor is also located on membranes of cells in the human brain. Interestingly, patients with AIDS show neuropsychological deficits. This embodiment of the invention could thus enter any cells that attach to HIV particles, but would be activated only when HIV protease is present.
Other portions of viral proteins could also be fused to the proricin variant. One of these is a largely deglycosylated xe2x80x9cgp 120 corexe2x80x9d consisting of a truncated version of the native gp120 with mutations in some of the loops. P. D. Kwong et al., supra; R. Wyatt et al., The Antigenic Structure of the HIV gp120 Envelope Glycoprotein, 393 Nature 705-711 (1998). Several antigenic residues in gp120 interact with the CD4 protein. R. Wyatt et al., supra. Peptides based on these data can be fused to the proricin variant. Also, a short segment of the V3 loop of the gp120 protein can bind to xe2x80x9csecondaryxe2x80x9d cell receptors. J. B. Ghiara et al., Crystal Structure of the Principal Neutralization Site of HIV-1,264 Science 82-85 (1994). This segment could be flanked by cysteine residues to form a disulfide bond in the genetically engineered variant and thereby stabilize the immunogenic secondary structure of the V3 loop segment. A. Zhang et al., A Disulfide-bound HIV-1 V3 Loop Sequence on the Surface of Human Rhinovirus 14 Induces Neutralizing Responses Against HIV-1,380 Biol. Chem. 365-374 (1999).
Immunogenic peptides that represent epitopes on the HIV gag protein might also serve as targeting molecules. J. M. Claverie et al., T-immunogenic Peptides Are Constituted of Rare Sequence Patterns. Use in the Identification of T Epitopes in the Human Immunodeficiency Virus Gag Protein, 18 J. Eur. Immunol. 1547-1553 (1988). The gag segment will direct an agent that contains it to be incorporated into the virus particle forming at the inner surface of the plasma membrane. C. T. Wang et al., Sequence Requirements for Incorporation of Human Immunodeficiency Virus Gag-2-galactosidase Fusion Proteins into Virus-like Particles, 59 J. Med. Virol. 180-188 (1999); M. Bryant and L. Ratner, Myristoylation-dependent Replication and Assembly of Human Immunodeficiency Virus 1, 87 Proc. Nat l Acad. Sci. USA 523-527 (1990). An appropriate segment may include the N-terminus of the Pr55gag protein, which is attached to myristic acid in vivo, thereby causing the myristylated protein to associate with the inner surface of the plasma membrane and the nascent virus particle.
Peptide targeting molecules can be coupled to the proricin variant by expression of a recombinant fusion protein, chemical synthesis and attachment, or any other suitable method known in the art. Such techniques for producing recombinant fusion proteins are well-known in the art, and are described generally in, e.g., J. Sambrook and D. Russell, Molecular Cloning: A Laboratory Manual (3d ed., 2001). Reagents useful in applying such techniques, such as restriction endonucleases and the like, are widely known in the art and commercially available from any of several vendors.
The activity of HIV protease is regulated such that it is minimal in the cytoplasm and most active inside the virus particle. J. Gatlin et al., supra. The activity in the cytoplasm may be sufficient to activate the proricin variant of this invention. This phenomenon of high activity in the viral particle, however, could be used to advantage to kill the cells upon infection, prior to integration of the viral DNA into the host DNA. By attaching fatty acid residues to the proricin variant, it would bind to the inner surface of the cell membrane and be incorporated into the budding virus particle just as fatty-acid-labeled viral proteins are directed to the membrane. M. Bryant and L. Ratner, supra. In this sense, the fatty acid is a targeting molecule. Hydrophobic groups other than fatty acids could provide the same function. Fatty acids could be incorporated by chemical treatment with fatty acyl halides or by coupling fatty acids to the proricin via carbodiimide chemistry. Upon secondary infection, when the virus containing the proricin variant fuses with the membrane of a new cell, it would release the HIV-protease-activated proricin variant into the cytoplasm, where the disulfide bond is reduced and the A chain is released, and thereby kills the cell before integration of the viral genome. It may be necessary to use an HIV-protease-cleavable linker that is not the optimal sequence for cleavage such that activation of the variant is promoted only within virus particles. A second advantage of using a fatty acylated proricin variant is that the fatty acids could promote cellular internalization of the therapeutic agent, V. Y. Alakhov et al., Increasing Cytostatic Effects of Ricin A Chain and Staphylococcus aureus Enterotoxin A through In Vitro Hydrophobization with Fatty Acid Residues, 12 Biotechnol. Appl. Biochem. 94-98 (1990), bypassing the complex internalization route for wild-type ricin involving the endoplasmic reticulum. This has been demonstrated with the ricin A chain. V. Y. Alakhov et al., supra.
Mannose residues on the ricin B chain may be required for the lectin binding activity. M. Westby et al., Preparation and Characterization of Recombinant Proricin Containing an Alternative Protease-sensitive Linker Sequence, 3 Bioconjug. Chem. 375-381 (1992). Therefore, a preferred embodiment of the invention involves making the proricin variant in a eukaryotic host, such as yeast, to provide the necessary glycosylation. The mannose residues are also targeting molecules and can direct native ricin to cells displaying mannose receptors, leading to cell intoxication. D. L. Newton et al., supra; B. M. Simmons et al., supra.
If one of the targeting molecules mentioned above is used, it may be possible to eliminate one or more of the galactose binding sites of the ricin B chain by point mutations, D. L. Newton et al., supra, or deletions. The mannose residues may also be eliminated by expression of a gene construct in a prokaryotic host or by enzymatic removal. However, at least one of the galactose binding sites appears to be required for effective cellular internalization of native ricin. D. L. Newton et al., supra. Leaving the galactose binding sites intact allows the possibility that nonspecific entry into cells would occur despite the incorporation of other targeting molecules. Intoxication, however, would still require that the cells expressed a protease that would cleave the modified proricin-based agent. The advantage of leaving one or both galactose binding sites intact is that the internalization of the agent into cells would be enhanced. Of course, adding fatty acid residues may allow direct cellular internalization of the agent and avoid the need for galactose binding sites.
Viral vectors could also be used to deliver a gene coding for this antiviral agent to human cells. HIV-based vectors capable of accomplishing this are known in the art. E.g., R. Zufferey et al., Self-inactivating Lentivirus Vector for Safe and Efficient In Vivo Gene Delivery, 72 J. Virol. 9873-9880 (1998); H. Miyoshi et al., Development of a Self-inactivating Lentivirus Vector, 72 J. Virol. 8150-8157 (1998). Such an HIV-based vector would have the nucleic acid coding for the antiviral agent of the present invention inserted into it. This vector containing the nucleic acid coding for the antiviral agent would then be administered to patients according to methods known in the art. A percentage of the cells in the body would become transduced with this construct and would thereafter continuously produce the antiviral agent. The antiviral agent would circulate in the blood, creating a condition similar to that created by injection of exogenous antiviral agent into the patient. The circulating antiviral agent could enter any HIV injectable cell, but would be activated and kill the cell only in the presence of HIV protease.
In still another embodiment of the invention, one of the termini of the ricin A chain is extended by means of making a fusion protein, then an adenine or adenine-like group is attached chemically. A cysteine residue may be incorporated into the extension to facilitate chemical attachment of the adenine group. The adenine-like group should not contain adenosine since the adenine moiety is likely to be cleaved from the ribose by the A chain. In this embodiment, the X (protease cleavable) linker is contained within the extension and the B chain is removed. The adenine residue would occupy the active site of the ricin A chain and inhibit activity. The extension linker would contain a cleavage site of an HIV protease. Upon cleavage, the linker/adenine would be separated and inhibition would be relieved, activating the ricin A chain. Adenine and pteroic acid, an adenine-like molecule, are known inhibitors of ricin. A. Pallanca et al., Uncompetitive Inhibition by Adenine of the RNA-N-glycosidase Activity of Ribosome-inactivating Proteins, 1384 Biochim. Biophys. Acta 277-284 (1998); J. D. Robertus et al., Structural Analysis of Ricin and Implications for Inhibitor Design, 34 Toxicon 1325-1334 (1996). It should be noted that there are two cysteine residues in the primary sequence of the ricin A chain that do not participate in a disulfide bond and are apparently not conserved among plant and bacterial ricin-like toxins. Y. Kitaoka, Involvement of the Amino Acids Outside the Active-site Cleft in the Catalysis of Ricin A Chain, 257 Eur. J. Biochem. 255-262 (1998). Thus, these two cysteine residues may not be essential to the activity of the ricin A chain. These cysteine residues could be mutated to other residues, leaving only the cysteine in the extension linker. 8-Adeninethiol could be coupled to this cysteine residue with, for example, a bismaleimide reagent. Pteroic acid could be coupled to an amine in the linker by carbodiimide chemistry. Other adenine-like molecules and attachment chemistries are possible and are well known in the art.
The gene for the proricin variant according to the present invention is preferably produced by amplifying the intron-free gene for preproricin in castor beans. The preproricin DNA sequence is known, e.g. F. I. Lamb et al., Nucleotide Sequence of Cloned cDNA Coding for Preproricin, 148 Eur. J. Biochem. 265-270 (1985); J. W. Tregear and L. M. Roberts, The Lectin Gene Family of Ricinus communis: Cloning of a Functional Ricin Gene and Three Lectin Pseudogenes, 18 Plant Mol. Biol. 515-525 (1992); K. C. Halling et al., Genomic Cloning and Characterization of a Ricin Gene from Ricinus communis, 13 Nucleic Acids Res. 8019-8033 (1985); L. M. Roberts et al., Molecular Cloning of Ricin, 7 Targeted Diagn. Ther. 81-97 (1992); JP 1985102188-A; U.S. Pat. No. 5,622,838. Methods for amplifying selected DNA segments, U.S. Pat. Nos. 4,683,195; 4,683,202; 4,800,159; 4,965,188; PCR Technology: Principles and Applications for DNA Amplification (H. Erlich ed., Stockton Press, New York, 1989); PCR Protocols: A guide to Methods and Applications is et al. eds, Academic Press, San Diego, Calif., 1990), and synthesizing oligonucleotides, S. A. Narang et al., 68 Meth. Enzymol. 90 (1979) (phosphotriester method); E. L. Brown et al., 68 Meth. Enzymol. 109 (1979) (phosphodiester method); U.S. Pat. Nos. 4,356,270; 4,458,066; 4,416,988; 4,293,652; N. D. Sinha et al., 24 Tetrahedron Lett. 5843 (1983); N. D. Sinha et al., 12 Nucl. Acids Res. 4539 (1984); N. D. Sinha et al., 15 Nucl. Acids Res. 397 (1987); N. D. Sinha et al., 16 Nucl. Acids Res. 319 (1988), are well known in the art. Methods for joining such amplified DNAs and synthetic DNAs are also well known in the art, e.g. J. Sambrook and D. Russell, Molecular Cloning: A Laboratory Manual (3d ed., 2001); T. Maniatis et al., Molecular Cloning: A Laboratory Manual (1982); F. Ausubel et al., Current Protocols in Molecular Biology (1987).
Next, the DNA encoding the preproricin variant containing the HIV protease site is cloned in an appropriate expression vector for expression in a suitable host cell, as is well known in the art. Methods for cloning in these vectors are well known in the art, e.g., Sambrook, supra; Maniatis, supra; Ausubel, supra. Such suitable host cells are preferably eukaryotic cells, but expression in prokaryotic cells is also intended to be within the scope of the present invention. Expression vectors for expression in eukaryotic cells, e.g., yeast, insect, and mammalian cells, and prokaryotic cells, e.g., bacterial cells, are well known in the art and are commercially available from numerous companies or from depositories such as the American Type Culture Collection. Such expression vectors generally contain the necessary promoters, transcription termination signals, translation initiation and termination signals, and the like for expression of the cloned DNA in the selected host cell. The cloned DNA is then expressed in a selected host transformed with the expression vector containing the variant preproricin DNA. Transformation of bacterial and eukaryotic cells is well known in the art and can be carried out by any appropriate method, such as calcium-dependent methods, heat shock, electroporation, and other effective methods and combinations thereof. The expressed protein is then purified, preferably by affinity of the ricin B chain for galactose-labeled agarose beads if the variant contains functional galactose binding sites of the B chain. Such affinity-based purification methods are also well known in the art.
Next, the purified proricin variant is tested for susceptibility to HIV protease and various other proteases to determine if the proricin variant is indeed selectively activated by HIV protease. As a first matter, the HIV protease should digest the proricin variant peptide, resulting in ricin A and B chains linked by a disulfide bond. The resulting ricin is chemically reduced and tested for ricin A chain activity against ribosomes using commercial in vitro translation kits.
The proricin variant containing the HIV protease site can be hydrophobized, as briefly described above. This modification has several advantages, including making it easier for the proricin variant to enter cells in vivo. Further, the hydrophobic agent coupled to the proricin variant has a tendency to direct the proricin variant to the inner surface of the cell membrane, as described above, where it can be incorporated into HIV particles during budding of the HIV particles through the cell membrane. Preferred hydrophobic agents include bile acids, sterols, and saturated and unsaturated fatty acids. Preferred bile acids include cholic acid, deoxycholic acid, chenodeoxycholic acid, lithocholic acid, ursocholic acid, ursodeoxycholic acid, isoursodeoxycholic acid, lagodeoxycholic acid, glycocholic acid, taurocholic acid, glycodeoxycholic acid, glycochenodeoxycholic acid, dehydrocholic acid, hyocholic acid, hyodeoxycholic acid, and mixtures thereof. Preferred sterols include cholestanol, coprostanol, cholesterol, epicholesterol, ergosterol, ergocalciferol, and mixtures thereof. Preferred saturated and unsaturated fatty acids comprise about 4 to 20 carbon atoms, such as butyric acid, valeric acid, caproic acid, caprylic acid, capric acid, lauric acid, myristic acid, palmitic acid, stearic acid, oleic acid, linoleic acid, linolenic acid, eleostearic acid, and mixtures thereof. The hydrophobic agent can be coupled to the proricin variant by chemical conjugation or, in appropriate cases, by using enzymes. For example, an alkanoic acid can be coupled to the proricin variant by forming an amide bond between the carboxylic acid group of the alkanoic acid and an amino group of the proricin variant. The amino group can be the N-terminal amino group of the proricin variant and/or one or more of the xcexc-amino groups of lysine residues. As is well known in the art, this reaction could be carried out chemically by activating the alkanoic acid by conversion to an acid chloride, and then reacting the acid chloride with the proricin variant to result in formation of the amide bond.
In still another embodiment of the present invention, a targeting moiety is coupled to the proricin variant. For example, a peptide targeting agent can be coupled to the proricin molecule either chemically or by recombinant DNA methodology. In some embodiments, the compositions are constructed by chemically conjugating the targeting moiety to the proricin variant. xe2x80x9cChemically conjugatingxe2x80x9d the targeting moiety to the proricin variant, as that term is used herein, means covalently bonding the targeting moiety to the proricin variant either directly or by way of a spacer moiety. Such spacer moieties can include heterobifunctional crosslinkers, such as are well known in the art. Peptide portions of the compositions of the present invention can be produced in a genetically engineered organism, such as E. coli or yeast, as a xe2x80x9cfusion protein.xe2x80x9d That is, a hybrid DNA containing a sequence of nucleotides encoding the targeting moiety and the proricin variant can be constructed by recombinant DNA technology. This hybrid DNA can be inserted into an organism such that the xe2x80x9cfusion proteinxe2x80x9d encoded by the hybrid DNA is expressed, as described above. The fusion protein can then be purified by standard methods, including affinity chromatography. If the targeting moiety is a relatively short peptide, such peptide can also be chemically synthesized. Methods for synthesis of peptides are well known in the art. E.g., R. B. Merrifield, Solid Phase Peptide Synthesis,32 Adv. Enzymol. Relat. Areas Mol. Biol. 221-96 (1969); R. B. Merrifield et al., 21 Biochemistry 5020-31 (1982) (solid phase peptide synthesis); Houghten, 82 Proc. Nat""l Acad. Sci. USA 5131-35 (1985) (solid phase peptide synthesis); Hunkapiller et al., 310 Nature 105-111 (1984).